400 bulletin: museum of comparative zoology. 



of the smallness of the type-specimen. Since the type of C. latinasus 

 is most probably lost, we consider it advisable to follow Boulenger's 

 suggestion and to place Colostethus in the synonymy of Phyllobates. 



The number of species of Phyllobates is greatly increased by the 

 referring of Phyllcdromus, Prostherapis, and Colostethus to its syn- 

 onymy. It is highly probable that some of the species are not actually 

 distinct from others. The name P. pvklicllus is based upon a figure 

 only (Espada, Vertebrados del Viaje al Pacifico, Batracios, 1875, pi. 3, 

 fig. 3, 3a, 3b, and 3c) of a frog, without data. Unfortunately the 

 figures do not show the ventral surface. Seven species of Phyllobates 

 are described with a dorsal pattern very similar to that represented 

 as P. i^nlchdhis. These seven species are most readily distinguishable 

 by their ventral coloration. Any one of them may be referable to 

 P. j^ulchellvs but without a description of their ventral surfaces it is 

 impossible to say which of them should bear the name. Our speci- 

 mens of P. infra gidfatns are very similar to Espada's figure. More- 

 over, we have recently examJned some specimens (U. S. N. M. 61,763, 

 61,764) of the same species from Huigra, Ecuador. These offer further 

 evidence of the wide distribution of the species in Ecuador and north- 

 ern Peru. It is probable that Espada's t^pe of P. jndchcllus came 

 from Ecuador. But until this type-specimen can be examined and 

 sufficiently characterized we have no right to refer Boulenger's P. 

 infragidfatus to that species. The name P. jmlchcUus must be dropped 

 until the type-specimen can be examined. 



We have remarked about the similarity of color-pattern in several 

 species of Phyllobates. The dorsal color-pattern shows little variation 

 in our enormous series of P. mfragiittahis. The ground-tone may be 

 light or dark grey. The pale specimens, m.ostly from the lowlands, 

 have the dorsal spotting very distinct and in sharp contrast to the 

 light grey ground-tone. Dark specimens are uniform above, the spots 

 being concealed by the ground-tone. The characteristic pair of spots 

 on the chest are indistinct in the darkest and practically wanting in 

 the very lightest specimens of the series. 



At Querocotilla one specimen was found which appeared nearly a 

 uniform pea-green. The specimen was put in a jar of fixing fluid 

 (.5% formol) and in the course of half an hour the green tones were 

 replaced by greys and the ordinary pattern appeared making the speci- 

 men indistinguishable from the others. 



The species of Phyllobates are all small, and possess few features 

 by which they may be distinguished from one another. Still it has 

 been considered advisable to attempt a key, unfortunately based 



